986431 - Investigate why tests are failing with "Keyboard not interpreted as displayed. Debug is_displayed(): False"

Status: RESOLVED WORKSFORME

(Firefox OS Graveyard :: Gaia::UI Tests, defect, P1)

Comment 1

[[:AndreiH]]

This fails on CI with:

Traceback (most recent call last):
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/.env/local/lib/python2.7/site-packages/marionette_client-0.7.5-py2.7.egg/marionette/marionette_test.py", line 163, in run
testMethod()
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/tests/python/gaia-ui-tests/gaiatest/tests/functional/clock/test_clock_create_new_alarm.py", line 35, in test_clock_create_new_alarm
new_alarm.type_alarm_label(alarm_label_text)
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/tests/python/gaia-ui-tests/gaiatest/apps/clock/regions/alarm.py", line 33, in type_alarm_label
self.keyboard.send(value)
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/tests/python/gaia-ui-tests/gaiatest/apps/keyboard/app.py", line 213, in send
self.switch_to_keyboard()
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/tests/python/gaia-ui-tests/gaiatest/apps/keyboard/app.py", line 149, in switch_to_keyboard
%keyboards.is_displayed())
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/tests/python/gaia-ui-tests/gaiatest/apps/base.py", line 54, in wait_for_condition
Wait(self.marionette, timeout).until(method, message=message)
File "/var/jenkins/workspace/b2g.hamachi.mozilla-central.ui/.env/local/lib/python2.7/site-packages/marionette_client-0.7.5-py2.7.egg/marionette/wait.py", line 143, in until
cause=last_exc)
TimeoutException: TimeoutException: Timed out after 30.0481901169 seconds with message: Keyboard not interpreted as displayed. Debug is_displayed(): False

I'll investigate this and try to come up with a fix, and start an adhoc job

Comment 2

[[:AndreiH]]

Adhoc : http://selenium.qa.mtv2.mozilla.com:8080/view/B2G%20Master/job/b2g.hamachi.mozilla-central.ui.adhoc/301/console

Comment 3

[[:AndreiH]]

I can see more tests failing for this reason: 
test_keyboard_predictive_key.py
test_clock_create_new_alarm.py
test_clock_add_alarm_multiple_times.py

Comment 4

[Jason Smith [:jsmith]]

Can we do exploratory testing on trunk to check if predictive text in the keyboard & the clock works fine?

Comment 5

[Viorela Ioia [:viorela]]

I was not able to reproduce this locally, even though I ran the tests several times. Manually neither. From the stacktrace it looks like the keyboard is not displayed.

Comment 6

[Florin Strugariu [:Bebe]]

Attachment: Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17647/files

Comment 7

[Florin Strugariu [:Bebe]]

http://selenium.qa.mtv2.mozilla.com:8080/view/B2G Hamachi/job/b2g.hamachi.mozilla-central.ui.adhoc/315

Comment 8

[[:AndreiH]]

test_clock_create_new_alarm seem to work just fine with this fix.
Let's test also the recently 2 fails:
test_clock_add_alarm_multiple_times
and
test_url_keyboard  from today's build.
Started adhoc for "test_url_keyboard":
http://selenium.qa.mtv2.mozilla.com:8080/view/B2G%20Master/job/b2g.hamachi.mozilla-central.ui.adhoc/316/console

Comment 9

[Zac C (:zac)]

Comment on attachment 8397056 [details]
Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17647/files

r- , I'd be extremely surprised if this resolved the problem because each time it times out we know this value is False already.

The problem is likely to lie in the step *before* this one, or the OOM killer killing the keyboard.

Comment 10

[Zac C (:zac)]

Or the keyboard crashing.

Comment 11

[Dave Hunt [:davehunt] [he/him] ⌚BST]

Comment on attachment 8397056 [details]
Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17647/files

I think you need to provide some more context to your fix. What investigation have you carried out, why is this an appropriate fix. Without that extra detail, I'm with Zac in feeling that this is a superfluous check.

Comment 12

[[:AndreiH]]

The latest failures when testing Keyboard in UI Tests:
http://selenium.qa.mtv2.mozilla.com:8080/job/b2g.hamachi.mozilla-central.ui.2/264/consoleFull

Comment 13

[Zac C (:zac)]

This has been stable lately, we didn't change the tests at all.

Comment 14

[Viorela Ioia [:viorela]]

This issue is reproducible in the latest master build. We have test_clock_create_new_alarm failing with this error: http://selenium.qa.mtv2.mozilla.com:8080/view/B2G%20Hamachi/job/b2g.hamachi.mozilla-central.ui.smoketest/728/testReport/junit/test_clock_create_new_alarm/TestClockCreateNewAlarm/test_clock_create_new_alarm/

Gaia      725a23802708eb70e3d7e8a2ce7179adbac806e4
Gecko     https://hg.mozilla.org/mozilla-central/rev/d7c07694f339
BuildID   20140429040202
Version   32.0a1
ro.build.version.incremental=eng.tclxa.20131223.163538
ro.build.date=Mon Dec 23 16:36:04 CST 2013

I was not able to reproduce it locally.

Comment 15

[[:AndreiH]]

This needs more investigation. Was reproduced in latest build on test_keyboard_basic (test_keyboard.TestKeyboard)

Build:
Gaia      15ac34804eb8b3c9b2582d7cf754c57e23182df6                         
Gecko     https://hg.mozilla.org/mozilla-central/rev/cf89b5d018f8          
BuildID   20140509040202                                                   
Version   32.0a1                                                           
ro.build.version.incremental=eng.tclxa.20131223.163538                     
ro.build.date=Mon Dec 23 16:36:04 CST 2013

Comment 16

[Viorela Ioia [:viorela]]

I've been able to reproduce this failure by running test_clock_create_new_alarm.py on desktopb2g locally: https://pastebin.mozilla.org/5331363
Also, I opened a pull request for trying to re-enable the clock tests on Travis and I saw them failing with the same error: https://travis-ci.org/mozilla-b2g/gaia/jobs/26662625#L1864
I'm still investigating this failure

Comment 17

[Zac C (:zac)]

Viorela, I think you need to look at the steps before the keyboard.

Possibly the step that taps onto the input field is not tapping in the right spot and thus the keyboard is never triggered to open.

Comment 18

[Florin Strugariu [:Bebe]]

we can see this reproducing in:

Gaia      fadfafa17f5175203b8b9457bfb95e5816f54f58
Gecko     https://hg.mozilla.org/mozilla-central/rev/b17cad2d1e5e
BuildID   20140729040211
Version   34.0a1
ro.build.version.incremental=109
ro.build.date=Mon Jun 16 16:51:29 CST 2014
B1TC00011220

Comment 19

[Viorela Ioia [:viorela]]

This is also reproducible in latest master build, test_cost_control_data_alert_mobile.py  is failing: http://jenkins1.qa.scl3.mozilla.com/job/flame.mozilla-central.ui.functional.non-smoke/127/HTML_Report/
It looks like the input field was not tapped, that's why the keyboard was not displayed. If I tap it manually, the test is passing.
I'll still work on this. 

Device firmware (date)     28 Aug 2014 08:29:26
Device firmware (incremental)     eng.cltbld.20140828.112916
Device firmware (release)     4.3
Device identifier     flame
Gaia date     01 Sep 2014 11:20:22
Gaia revision     44bf2e3bc5dd
Gecko build     20140902040205
Gecko revision     c360f3d1c00d
Gecko version     34.0a1

Comment 20

[Tony Chung [:tchung]]

:geo, are you taking this bug?  this one's hot and being requested from our partners, so gaining traction on this.

Comment 21

[Geo Mealer [:geo] -- This account is inactive after 2015-07-07]

(In reply to Tony Chung [:tchung] from comment #20)
> :geo, are you taking this bug?  this one's hot and being requested from our
> partners, so gaining traction on this.

Yes. I'll assign it to myself as soon as I start investigation on it.

Comment 22

[Martijn Wargers (dead)]

I tried to reproduce this on desktop B2G on Mac, but I wasn't able to.

Comment 23

[Geo Mealer [:geo] -- This account is inactive after 2015-07-07]

Yeah, I don't think anyone has been able to reproduce this recently. I haven't had any luck either, and Viorela mentioned not being able to do so anymore either.

We discussed in the last meeting setting this to WORKSFORME and re-opening if we have another example case. I'll leave this comment here for a day or two, then I'm going to do that.

Comment 24

[Geo Mealer [:geo] -- This account is inactive after 2015-07-07]

NI'ing myself to remind myself to do this.

Comment 25

[Geo Mealer [:geo] -- This account is inactive after 2015-07-07]

Per inability to repro mentioned in comment 23, marking this WORKSFORME. If it reappears, please reopen this bug with as much information for reproduction as possible.